Senior Machine Learning Engineer, Agentic Context Retrieval and Search (Nova)

Royper Royco LLC; United States; Remote; USD 180,000 - 240,000; Full time

Job summary

Iterable is seeking a Senior Machine Learning Engineer to lead context retrieval and search for Nova, the intelligence layer. You will build the retrieval stack and search subagents that power grounded, context-aware agent responses.

You will have autonomy from research to production, shaping the direction of AI at Iterable while mentoring teammates and driving high-impact improvements across the system.

Qualifications

• Must-have experience building retrieval systems for structured and unstructured data.
• Proficiency in semantic query understanding and multi-turn workflows.
• Experience delivering production-grade ML systems and mentoring engineers.

Responsibilities

• Build retrieval across structured data (tables, SQL, dashboards, code) and unstructured content.
• Develop semantic query understanding and entity resolution for agents.
• Create pipelines to extract entities, embeddings, summaries, and metadata from data.
• Develop search subagents that decide what to retrieve and perform multi-hop searches.
• Measure retrieval quality with ranking metrics and online tests; iterate accordingly.
• Mentor engineers and influence the technical direction for Nova.

Skills

Retrieval systems

Semantic understanding

Entity resolution

Multi-hop search

ML engineering

Mentorship

Job description

Iterable is the AI customer engagement platform, built for enterprise scale, loved by teams, and trusted by global brands like Redfin, SeatGeek, Priceline, Calm, and Box. Our platform empowers organizations to activate customer data from any source, design seamless cross-channel experiences, and optimize engagement with Nova Intelligence, the AI layer of our platform, all with enterprise-grade security and compliance. Today, nearly 1,200 brands across 50+ countries rely on Iterable to drive growth, deepen customer relationships, and deliver joyful customer experiences.

Our success is powered by extraordinary people who bring our core values to life every day: Be an Owner, Growth Mindset, Run as One, and Transparency. And the market has noticed: G2 recognizes Iterable as a Leader globally, spanning the Americas, EMEA, APAC, ANZ, and Latin America, with category leadership in Personalization, Marketing Automation, Push Notifications, and Mobile Marketing, and a 4.5/5-star rating from hundreds of users.

Gartner named Iterable a Challenger in the Gartner Magic Quadrant for Multichannel Marketing in both 2024 and 2025. We've also earned TrustRadius Top Rated honors in 2024 and 2025, were named a TrustRadius Buyer's Choice for 2026, and were featured in Snowflake's Modern Marketing Data Stack 2026.

With a global presence that includes offices in San Francisco, Denver, London, Sydney, and Lisbon, plus remote employees worldwide, we are committed to building a diverse and inclusive workplace. We welcome candidates from all backgrounds and encourage you to apply. Learn more about our story and mission on our Culture and About Us pages. let's shape the future of customer engagement together!

Location: US

About the role

As a high-tech marketing platform, Iterable works with thousands of large-scale and diverse enterprise customers. Iterable Nova is our intelligence layer. It helps customer experience teams adapt in real time and improve every interaction.

As a Senior Machine Learning Engineer on Nova, you'll lead context retrieval and search. The goal is agent responses that are grounded in evidence, aware of the customer's context, and based on what a query means rather than its literal words. You will build the retrieval stack and the search subagents that run on top of it.

You will have real autonomy: from research and experimentation through to systems running reliably in production, you'll own this work and help shape the direction of AI at Iterable.

What You'll Do

• Build retrieval across structured data (tables, SQL, dashboards, code) and unstructured content (docs, tickets, chat, media), using each source's own signals.
• Build semantic query understanding for agents: query rewriting, decomposition, intent classification, and entity resolution for multi-turn workflows.
• Build pipelines that extract entities, embeddings, summaries, and metadata from customer data, and keep them fresh.
• Build search subagents that decide what to retrieve, run multi-hop searches, check whether results are good enough, and flag when retrieval fails.
• Measure quality with ranking metrics (nDCG, MRR, Recall@K), LLM judges, human labels, and online tests. Track how retrieval affects final responses, and use the results to improve both retrieval and the harness.
• Mentor engineers on the team and help set the technical direction for Nova.